In a network communication represented by a Internet Protocol (IP) network, it is supposed that an address (IP address in the IP network) that is not overlapped is allocated to respective terminal apparatuses.
When the IP address is overlapped, the terminal apparatus to which the overlapped IP address is allocated cannot normally make communication. An address resolution protocol (ARP) has been known as a technique of detecting the overlap of the address (e.g., see Japanese Laid-open Patent Publication No. 5-56050).
Specifically, in the ARP, a broadcast packet is transmitted so as to send the originator IP address to all terminal apparatuses. When the transmitted packet reaches the terminal apparatus having the overlapped IP address, the overlap of the IP address can be detected.
When data necessary for update is acquired through the IP network in case where firmware data of the terminal apparatus connected to the IP network is updated, it is necessary to assign the IP address to the terminal apparatus that is the subject to be updated. The assignment of the IP address is performed such that an operator firstly selects a usable IP address, and then, manually inputs the selected IP address to the subject apparatus whose firmware data is to be updated.
However, when an error is caused in the selection of the IP address or in the input, the IP address of the terminal apparatus that is the subject to be updated and the IP address of the other terminal apparatus that has already been installed might be overlapped.
When a new terminal apparatus is added to make an IP communication with a specific device in the state in which all IP addresses in a segment have already been assigned, the IP address might be overlapped with the IP address of any one of the terminal apparatuses in the segment.
There is also an environment in which the IP address is not managed. When the IP communication is needed in the environment in which the IP address is not managed, an optional IP address, which is considered not to be used, is selected and assigned to the address of the own terminal. However, there may be the case in which the selected address has already been used in the other terminal apparatus.
Whether an IP address has already been used or not is generally determined whether there is a response in Ping or not. However, a device that does not respond to Ping increases due to the implementation of firewall software to a computer system.
Therefore, even the selected IP address that is considered not to be used because of no response to the Ping might be overlapped.
The IP communication is sometimes demanded between specific terminal apparatuses, even when the IP addresses might be overlapped as described above. There has been proposed a method, as a method of the IP communication allowing the overlap of the IP address, in which a proxy response apparatus is added between network, wherein identification information of a slave device is converted into an IP address by the proxy response apparatus, and an IP packet is transmitted to another network (see, for example, Japanese Laid-open Patent Publication No. 9-331325).
However, in a conventional technique, a normal communication cannot be made, when addresses are overlapped in a segment. For example, even when a proxy response apparatus is used, an IP address that is not overlapped has to be assigned to a connection part between the proxy response apparatus and the network of a communication destination. There is also a problem that an apparatus for performing a proxy response is needed.
In view of this, an important problem is the implementation of a control method in which an IP communication can be made between specific terminal apparatuses without giving influence to communication to other terminal apparatus, even if an overlap is caused due to an erroneous selection or erroneous input of an address.